Vertical 9:16 video gets the headlines, but the scrolling feed is still where a huge share of social watch time happens — and the feed is not a 9:16 surface. Instagram's in-grid player, LinkedIn's timeline, X's timeline, and Facebook's Feed all present video inside a scrolling column where square 1:1 and portrait 4:5 own the real estate. Drop a 16:9 clip into those placements and it renders as a short letterboxed strip. Deliver the same edit at 4:5 and it fills the column edge to edge, occupying more than twice the screen height of the widescreen version. This guide covers when 1:1 and 4:5 are the right delivery shapes, how composing for a feed differs from composing for full-screen vertical, and the exact workflow for converting a 16:9 master into square and 4:5 cuts inside Skrrol's browser studio — including the reframe modes that keep your subject in the shot instead of on the cutting-room floor.
Why 1:1 and 4:5 still matter in a 9:16 world
The short-form boom trained everyone to think social video means full-screen vertical, and for TikTok, Reels, Shorts, and Stories it does. But those are immersive surfaces — the video takes over the display and the interface floats on top. Feeds work differently. A feed is a column the user scrolls with a thumb, and video sits inline between text posts, photos, and ads. On an inline surface, the shape of your file directly controls how much screen you occupy while the viewer decides whether to stop. That is the whole argument for 1:1 and 4:5 in one sentence: taller-than-wide files buy more pixels in the feed.
The platform specs make it concrete. Instagram's feed recommends 4:5 at 1080×1350 for maximum visual real estate, supports 1:1 at 1080×1080, and renders 16:9 at just 1080×608. LinkedIn runs 16:9 as its standard with 1:1 for the feed. X prefers 16:9 but supports 1:1. Facebook's Feed takes 16:9 with 1:1 as the legacy in-grid shape, and Pinterest supports 1:1 alongside its preferred 9:16. Even TikTok accepts square files — it just pillarboxes them in its full-screen player, which is exactly why square belongs in feeds and not on full-screen surfaces. Run the arithmetic on Instagram's numbers and the case closes itself: at the same column width, a 4:5 file shows roughly 2.2 times the pixels of a 16:9 file, and a 1:1 file about 1.8 times. Same edit, same slot, double the presence.
So the practical delivery matrix for most creators ends up being three shapes: 16:9 for YouTube and other landscape surfaces, 9:16 for the full-screen short-form players, and 1:1 or 4:5 for the feeds. The master is usually 16:9. Everything else is a conversion.
How feed framing differs from full-screen vertical
Reframing for 9:16 and reframing for 1:1 or 4:5 are different problems, and treating them as the same job is how subjects end up half out of frame. Start with the geometry. Cropping a 1920×1080 source to 9:16 keeps a 608-pixel-wide slice — under a third of the original width. Cropping to 1:1 keeps 1080 pixels, just over half, and 4:5 keeps 864, about 45 percent. Square and 4:5 crops are therefore far more forgiving than vertical ones: framing that cannot survive a 9:16 crop often survives a square one intact. A two-person interview that will never fit a vertical frame usually fits a square frame comfortably.
Composition weight shifts too. Full-screen vertical is a top-and-bottom problem — you keep critical content out of the lower third where captions and platform controls sit, and hooks run high in the frame. Feed formats are center-weighted: the eye lands in the middle of the tile as it scrolls past, no interface floats over your pixels, and the frame reads more like a photograph. Subjects belong at or near center, faces need to be large enough to register in a small tile, and edge detail is expendable because the edges are what the crop discards. And since feeds autoplay muted for most viewers, burned-in captions stay non-negotiable — just place them inside the safe area of your smallest target shape rather than parking them at the very bottom of the frame.
Converting 16:9 to 1:1 and 4:5, step by step
Skrrol's aspect ratio converter treats a format change as a project-level switch plus a per-clip framing decision, and the whole conversion runs in the browser with no upload. Six steps take a 16:9 edit to a finished square or 4:5 deliverable.
- Set the project aspect. Open project settings and pick the target — 1:1 or 4:5 here, with 16:9, 9:16, and 21:9 also on the list. The canvas reshapes to match.
- Choose a reframe mode per clip. Letterbox adds black bars and preserves everything, Center Crop chops the edges and keeps the middle, Manual sets a custom crop, and Subject Track has the AI follow the action.
- Preview the result. Scrub the timeline to see how each clip lands in the new aspect, and adjust the crop or the subject focus per clip where needed.
- Refine the subject path. If subject tracking misses a beat, drop manual keyframes on the clip's crop position to override the AI in just those frames.
- Export to the target aspect. The render writes the file at the target's native resolution — 1080×1080 for square — with the correct metadata baked in.
- Re-export for additional aspects. Duplicate the project, change the aspect, and re-export. The source media stays the same; only the framing changes.
The per-clip part matters more than it sounds. A three-minute edit might hold forty clips, and one reframe strategy almost never fits all of them: the locked-off product shot wants a center crop, the walking interview wants subject tracking, and the full-width title card wants letterboxing or a rebuild. Because the mode is chosen clip by clip, you make that call where it belongs instead of accepting one compromise across the whole timeline.
Choosing the right reframe mode
Letterbox preserves every pixel by fitting the full source inside the new canvas and filling the rest with black bars. It is the right call when the full width is genuinely load-bearing — screen recordings, wide graphics, archival footage you are not allowed to crop. In a feed, though, letterboxing spends the exact real estate you converted to gain, so treat it as the exception rather than the default.
Center Crop keeps the middle of the frame and discards the edges. For locked-off shots with a centered subject — talking heads, product-on-table, cooking top-downs — it converts a clip in seconds and looks deliberate. Manual covers the shots Center Crop almost fits: the subject is static but sits off-center, so you set the crop to a fixed offset position and leave it there.
Subject Track is the headline mode. The AI follows the person or product through the shot and shifts the crop to keep them centered as the canvas shape changes, so a horizontal clip where the speaker drifts across frame becomes a square clip where the speaker never leaves it. The tracker runs on the same on-device AI model that powers Skrrol's Smart Cutout, which keeps the reframe smooth instead of jittery, and it handles most single-subject shots well. Multi-subject scenes — two people trading lines, a hand and a product moving independently — force the tracker to pick, so use manual keyframes for precision there.
Composition tips for center-weighted frames
A few working habits make center-weighted delivery consistently better. First, frame for the crop at the shoot when you know feed cuts are coming: keep the subject inside the middle half of the 16:9 frame and treat the outer edges as bleed. Footage shot this way survives 1:1 nearly untouched and 4:5 with room to spare.
Second, watch the headroom. Square and 4:5 frames are taller relative to their width than 16:9, so the crop keeps more vertical space around a face — an interview framed with tight 16:9 headroom often ends up with the eyes sitting too low after conversion. Nudge the crop upward in Manual mode, or adjust the subject focus, until the eyes land around the upper third of the new frame.
Third, respect the edges. Lower-third graphics, tickers, and anything hugging the left or right border of a 16:9 frame sits in the discard zone of a 4:5 crop, which keeps only about 45 percent of the source width. Scrub the preview on every clip that carries text; where a graphic gets clipped, either letterbox that clip or rebuild the text at the new aspect inside the studio. Fourth, favor tighter shot sizes. Wide establishing shots lose their sweep in a small tile, while mediums and close-ups gain intimacy — in a feed, a face filling a square frame has a directness a landscape wide never achieves in the same slot.
Batch thinking: one edit, every format
The efficient multi-format workflow is to cut the story once and derive every delivery shape from that master, rather than editing per platform. In Skrrol the pattern is explicit: finish the 16:9 edit, duplicate the project, switch the duplicate's aspect to 1:1, walk the timeline setting reframe modes, export — then repeat for 4:5 and 9:16. The source media stays the same across every duplicate and only the framing changes, so each additional format costs minutes of reframe decisions, not a re-edit.
Sequence the passes so decisions carry forward. Do 16:9 to 1:1 first, because square is the most forgiving conversion and it forces you to catalog which clips need special handling. The 4:5 pass then inherits that thinking: every clip that survived square survives 4:5 with a slightly tighter side crop, and the clips you flagged for manual keyframes in square are the same ones to re-check. Audio needs no attention in any pass — aspect changes touch only the picture, so your mix, music, and loudness targets carry through untouched.
Common pitfalls when converting to 1:1 and 4:5
The first pitfall is stretching — resizing the picture to fit the new shape instead of reframing it. Nothing says amateur faster than a squeezed face. Skrrol's converter simply will not do it; every mode crops or letterboxes, so distortion is off the table. The second is shipping a fake conversion: a 16:9 video centered on a square canvas with bars baked in above and below. The platform treats that file as square, the bars upload as pixels, and your feed real estate is spent on black. A real conversion writes a native file — a full 1080×1080 of picture for square — which is exactly what the converter's native-resolution export produces.
The third is metadata. A file whose orientation metadata is wrong or missing can display incorrectly once uploaded; Skrrol writes the correct metadata for the target aspect at export so platforms recognize the shape without guessing. The fourth is starting from too little resolution. A 1:1 crop of a 1920×1080 source yields exactly 1080×1080 — a clean native square with no upscaling. A 4:5 crop of the same source is an 864×1080 slice, short of Instagram's 1080×1350 spec, so it has to scale up to reach it. A modest upscale is survivable, but the sharper path is a 4K master, where both crops land above delivery resolution with pixels to spare.
The last pitfall is trusting the conversion blind. Scrub the converted timeline end to end before export — the preview step exists precisely so a mis-tracked clip or a clipped graphic gets caught in review, not in the comments.
Where to start
If a 16:9 edit is sitting in your studio right now, the fastest way to internalize all of this is one real conversion. Duplicate the project, switch it to 1:1, set Subject Track on the talking-head clips and Center Crop on the locked-off ones, scrub the timeline, export, and watch the result next to the original. The second pass to 4:5 will take half as long as the first. After that, feed formats stop being an afterthought and become a standard final step of every delivery — one edit in, a full set of shapes out.